Ranjit Kumar - Research Methodology

aspects: The class, families living in the city or electorates from which you select you select your sample are called the population or study population, and are usually denoted by the letter N. The small group of students, families or electors from whom you collect the required information to estimate the average age of the class, average income or the election outcome is called the sample. The number of students, families or electors from whom you obtain the required information is called the sample size and is usually denoted by the letter n. The way you select students, families or electors is called the sampling design or sampling strategy. Each student, family or elector that becomes the basis for selecting your sample is called the sampling unit or sampling element. A list identifying each student, family or elector in the study population is called the sampling frame. If all elements in a sampling population cannot be individually identified, you cannot have a sampling frame for that study population. Your findings based on the information obtained from your respondents (sample) are called sample statistics. Your sample statistics become the basis of estimating the prevalence of the above characteristics in the study population. Your main aim is to find answers to your research questions in the study population, not in the sample you collected information from. From sample statistics we make an estimate of the answers to our research questions in the study population. The estimates arrived at from sample statistics are called population parameters or the population mean. Principles of sampling The theory of sampling is guided by three principles. To effectively explain these, we will take an extremely simple example. Suppose there are four individuals A, B, C and D. Further suppose that A is 18 years of age, B is 20, C is 23 and D is 25. As you know their ages, you can find out (calculate) their average age by simply adding 18 + 20 + 23 + 25 = 86 and dividing by 4. This gives the average (mean) age of A, B, C and D as 21.5 years. Now let us suppose that you want to select a sample of two individuals to make an estimate of the average age of the four individuals. To select an unbiased sample, we need to make sure that each unit has an equal and independent chance of selection in the sample. Randomisation is a process that enables you to achieve this. In order to achieve randomisation we use the theory of probability in forming pairs which will provide us with six possible combinations of two: A and B; A and C; A and D; B and C; B and D; and C and D. Let us take each of these pairs to calculate the average age of the sample: 1. A + B = 18 + 20 = 38/2 = 19.0 years; 2. A + C = 18 + 23 = 41/2 = 20.5 years; 3. A + D = 18 + 25 = 43/2 = 21.5 years; 4. B + C = 20 + 23 = 43/2 = 21.5 years;

5. B + D = 20 + 25 = 45/2 = 22.5 years; 6. C + D = 23 + 25 = 48/2 = 24.0 years. Notice that in most cases the average age calculated on the basis of these samples of two (sample statistics) is different. Now compare these sample statistics with the average of all four individuals – the population mean (population parameter) of 21.5 years. Out of a total of six possible sample combinations, only in the case of two is there no difference between the sample statistics and the population mean. Where there is a difference, this is attributed to the sample and is known as sampling error. Again, the size of the sampling error varies markedly. Let us consider the difference in the sample statistics and the population mean for each of the six samples (Table 12.1). TABLE 12.1 The difference between sample statistics and the population mean This analysis suggests a very important principle of sampling: Principle 1 – in a majority of cases of sampling there will be a difference between the sample statistics and the true population mean, which is attributable to the selection of the units in the sample. To understand the second principle, let us continue with the above example, but instead of a sample of two individuals we take a sample of three. There are four possible combinations of three that can be drawn: 1. 1 A + B + C = 18 + 20 + 23 = 61/3 = 20.33 years; 2. 2 A + B + D = 18 + 20 + 25 = 63/3 = 21.00 years; 3. 3 A + C + D = 18 + 23 + 25 = 66/3 = 22.00 years; 4. 4 B + C + D = 20 + 23 + 25 = 68/3 = 22.67 years. Now, let us compare the difference between the sample statistics and the population mean (Table 12.2). TABLE 12.2 The difference between a sample and a population average Compare the differences calculated in Table 12.1 and Table 12.2. In Table 12.1 the difference

between the sample statistics and the population mean lies between –2.5 and +2.5 years, whereas in the second it is between –1.17 and +1.17 years. The gap between the sample statistics and the population mean is reduced in Table 12.2. This reduction is attributed to the increase in the sample size. This, therefore, leads to the second principle: Principle 2 – the greater the sample size, the more accurate the estimate of the true population mean. The third principle of sampling is particularly important as a number of sampling strategies, such as stratified and cluster sampling, are based on it. To understand this principle, let us continue with the same example but use slightly different data. Suppose the ages of four individuals are markedly different: A = 18, B = 26, C = 32 and D = 40. In other words, we are visualising a population where the individuals with respect to age – the variable we are interested in – are markedly different. Let us follow the same procedure, selecting samples of two individuals at a time and then three. If we work through the same procedures (described above) we will find that the difference in the average age in the case of samples of two ranges between –7.00 and + 7.00 years and in the case of the sample of three ranges between –3.67 and +3.67. In both cases the range of the difference is greater than previously calculated. This is attributable to the greater difference in the ages of the four individuals – the sampling population. In other words, the sampling population is more heterogeneous (varied or diverse) in regard to age. Principle 3 – the greater the difference in the variable under study in a population for a given sample size, the greater the difference between the sample statistics and the true population mean. These principles are crucial to keep in mind when you are determining the sample size needed for a particular level of accuracy, and in selecting the sampling strategy best suited to your study. Factors affecting the inferences drawn from a sample The above principles suggest that two factors may influence the degree of certainty about the inferences drawn from a sample: 1. The size of the sample – Findings based upon larger samples have more certainty than those based on smaller ones. As a rule, the larger the sample size, the more accurate the findings. 2. The extent of variation in the sampling population – The greater the variation in the study population with respect to the characteristics under study, for a given sample size, the greater the uncertainty. (In technical terms, the greater the standard deviation, the higher the standard error for a given sample size in your estimates.) If a population is homogeneous (uniform or similar) with respect to the characteristics under study, a small sample can provide a reasonably good estimate, but if it is heterogeneous (dissimilar or diversified), you need to select a larger sample

to obtain the same level of accuracy. Of course, if all the elements in a population are identical, then the selection of even one will provide an absolutely accurate estimate. As a rule, the higher the variation with respect to the characteristics under study in the study population, the greater the uncertainty for a given sample size. Aims in selecting a sample When you select a sample in quantitative studies you are primarily aiming to achieve maximum precision in your estimates within a given sample size, and avoid bias in the selection of your sample. Bias in the selection of a sample can occur if: sampling is done by a non-random method – that is, if the selection is consciously or unconsciously influenced by human choice; the sampling frame – list, index or other population records – which serves as the basis of selection, does not cover the sampling population accurately and completely; a section of a sampling population is impossible to find or refuses to co-operate. Types of sampling The various sampling strategies in quantitative research can be categorised as follows (Figure 12.2):

FIGURE 12.2 Types of sampling in quantitative research random/probability sampling designs; non-random/non-probability sampling designs selecting a predetermined sample size; ‘mixed’ sampling design. To understand these designs, we will discuss each type individually. Random/probability sampling designs For a design to be called random sampling or probability sampling, it is imperative that each element in the population has an equal and independent chance of selection in the sample. Equal implies that the probability of selection of each element in the population is the same; that is, the choice of an element in the sample is not influenced by other considerations such as personal preference. The concept of independence means that the choice of one element is not dependent upon the choice of another element in the sampling; that is, the selection or rejection of one element does not affect the inclusion or exclusion of another. To explain these concepts let us return to our example

of the class. Suppose there are 80 students in the class. Assume 20 of these refuse to participate in your study. You want the entire population of 80 students in your study but, as 20 refuse to participate, you can only use a sample of 60 students. The 20 students who refuse to participate could have strong feelings about the issues you wish to explore, but your findings will not reflect their opinions. Their exclusion from your study means that each of the 80 students does not have an equal chance of selection. Therefore, your sample does not represent the total class. The same could apply to a community. In a community, in addition to the refusal to participate, let us assume that you are unable to identify all the residents living in the community. If a significant proportion of people cannot be included in the sampling population because they either cannot be identified or refuse to participate, then any sample drawn will not give each element in the sampling population an equal chance of being selected in the sample. Hence, the sample will not be representative of the total community. To understand the concept of an independent chance of selection, let us assume that there are five students in the class who are extremely close friends. If one of them is selected but refuses to participate because the other four are not chosen, and you are therefore forced to select either the five or none, then your sample will not be considered an independent sample since the selection of one is dependent upon the selection of others. The same could happen in the community where a small group says that either all of them or none of them will participate in the study. In these situations where you are forced either to include or to exclude a part of the sampling population, the sample is not considered to be independent, and hence is not representative of the sampling population. However, if the number of refusals is fairly small, in practical terms, it should not make the sample non- representative. In practice there are always some people who do not want to participate in the study but you only need to worry if the number is significantly large. A sample can only be considered a random/probability sample (and therefore representative of the population under study) if both these conditions are met. Otherwise, bias can be introduced into the study. There are two main advantages of random/probability samples: 1. As they represent the total sampling population, the inferences drawn from such samples can be generalised to the total sampling population. 2. Some statistical tests based upon the theory of probability can be applied only to data collected from random samples. Some of these tests are important for establishing conclusive correlations. Methods of drawing a random sample Of the methods that you can adopt to select a random sample the three most common are: 1. The fishbowl draw – if your total population is small, an easy procedure is to number each element using separate slips of paper for each element, put all the slips into a box and then pick them out one by one without looking, until the number of slips selected equals the sample size you decided upon. This method is used in some lotteries. 2. Computer program – there are a number of programs that can help you to select a random

sample. 3. A table of randomly generated numbers – most books on research methodology and statistics include a table of randomly generated numbers in their appendices (see, e.g., Table 12.3). You can select your sample using these tables according to the procedure described in Figure 12.3. The procedure for selecting a sample using a table of random numbers is as follows: Let us take an example to illustrate the use of Table 12.3 for random numbers. Let us assume that your sampling population consists of 256 individuals. Number each individual from 1 to 256. Randomly select the starting page, set of column (1 to 10) or row from the table and then identify three columns or rows of numbers. Suppose you identify the ninth column of numbers and the last three digits of this column (underlined). Assume that you are selecting 10 per cent of the total population as your sample (25 elements). Let us go through the numbers underlined in the ninth set of columns. The first number is 049 which is below 256 (total population); hence, the 49th element becomes a part of your sample. The second number, 319, is more than the total elements in your population (256); hence, you cannot accept the 319th element in the sample. The same applies to the next element, 758, and indeed the next five elements, 589, 507, 483, 487 and 540. After 540 is 232, and as this number is within the sampling frame, it can be accepted as a part of the sample. Similarly, if you follow down the same three digits in the same column, you select 052, 029, 065, 246 and 161, before you come to the element 029 again. As the 29th element has already been selected, go to the next number, and so on until 25 elements have been chosen. Once you have reached the end of a column, you can either move to the next set of columns or randomly select another one in order to continue the process of selection. For example, the 25 elements shown in Table 12.4 are selected from the ninth, tenth and second columns of Table 12.3. TABLE 12.3 Selecting a sample using a table for random numbers

Source: Statistical Tables, 3e, by F. James Rohlf and Robert R. Sokal. Copyright © 1969, 1981, 1994 by W.H. Freeman and Company. Used with permission. FIGURE 12.3 The procedure for using a table of random numbers TABLE 12.4 Selected elements using the table of random numbers

Sampling with or without replacement Random sampling can be selected using two different systems: 1. sampling without replacement; 2. sampling with replacement. Suppose you want to select a sample of 20 students out of a total of 80. The first student is selected out of the total class, and so the probability of selection for the first student is 1/80. When you select the second student there are only 79 left in the class and the probability of selection for the second student is not 1/80 but 1/79. The probability of selecting the next student is 1/78. By the time you select the 20th student, the probability of his/her selection is 1/61. This type of sampling is called sampling without replacement. But this is contrary to our basic definition of randomisation; that is, each element has an equal and independent chance of selection. In the second system, called sampling with replacement, the selected element is replaced in the sampling population and if it is selected again, it is discarded and the next one is selected. If the sampling population is fairly large, the probability of selecting the same element twice is fairly remote. FIGURE 12.4 The procedure for selecting a simple random sample Specific random/probability sampling designs There are three commonly used types of random sampling design. 1. Simple random sampling (SRS) – The most commonly used method of selecting a probability sample. In line with the definition of randomisation, whereby each element in the population is given an equal and independent chance of selection, a simple random sample is selected by the procedure presented in Figure 12.4. To illustrate, let us again take our example of the class. There are 80 students in the class, and so the first step is to identify each student by a number from 1 to 80. Suppose you decide to select a sample of 20 using the simple random sampling technique. Use the fishbowl draw, the table for random numbers or a computer program to select the 20 students. These 20 students

become the basis of your enquiry. 2. Stratified random sampling – As discussed, the accuracy of your estimate largely depends on the extent of variability or heterogeneity of the study population with respect to the characteristics that have a strong correlation with what you are trying to ascertain (Principle 3). It follows, therefore, that if the heterogeneity in the population can be reduced by some means for a given sample size you can achieve greater accuracy in your estimate. Stratified random sampling is based upon this logic. In stratified random sampling the researcher attempts to stratify the population in such a way that the population within a stratum is homogeneous with respect to the characteristic on the basis of which it is being stratified. It is important that the characteristics chosen as the basis of stratification are clearly identifiable in the study population. For example, it is much easier to stratify a population on the basis of gender than on the basis of age, income or attitude. It is also important for the characteristic that becomes the basis of stratification to be related to the main variable that you are exploring. Once the sampling population has been separated into non- overlapping groups, you select the required number of elements from each stratum, using the simple random sampling technique. There are two types of stratified sampling: proportionate stratified sampling and disproportionate stratified sampling. With proportionate stratified sampling, the number of elements from each stratum in relation to its proportion in the total population is selected, whereas in disproportionate stratified sampling, consideration is not given to the size of the stratum. The procedure for selecting a stratified sample is schematically presented in Figure 12.5. 3. Cluster sampling – Simple random and stratified sampling techniques are based on a researcher’s ability to identify each element in a population. It is easy to do this if the total sampling population is small, but if the population is large, as in the case of a city, state or country, it becomes difficult and expensive to identify each sampling unit. In such cases the use of cluster sampling is more appropriate. Cluster sampling is based on the ability of the researcher to divide the sampling population into groups (based upon visible or easily identifiable characteristics), called clusters, and then to select elements within each cluster, using the SRS technique. Clusters can be formed on the basis of geographical proximity or a common characteristic that has a correlation with the main variable of the study (as in stratified sampling). Depending on the level of clustering, sometimes sampling may be done at different levels. These levels constitute the different stages (single, double or multiple) of clustering, which will be explained later. Imagine you want to investigate the attitude of post-secondary students in Australia towards problems in higher education in the country. Higher education institutions are in every state and territory of Australia. In addition, there are different types of institutions, for example universities, universities of technology, colleges of advanced education and colleges of technical and further education (TAFE) (Figure 12.6). Within each institution various courses are offered at both undergraduate and postgraduate levels. Each academic course could take three to four years. You can imagine the magnitude of the task. In such situations cluster sampling is extremely useful in selecting a random sample. The first level of cluster sampling could be at the state or territory level. Clusters could be grouped according to similar characteristics that ensure their comparability in terms of student population. If this is not easy, you may decide to select all the states and territories and then select a sample at the institutional level. For example, with a simple random technique, one

institution from each category within each state could be selected (one university, one university of technology and one TAFE college). This is based upon the assumption that institutions within a category are fairly similar with regards to student profile. Then, within an institution on a random basis, one or more academic programmes could be selected, depending on resources. Within each study programme selected, students studying in a particular year could then be selected. Further, selection of a proportion of students studying in a particular year could then be made using the SRS technique. The process of selecting a sample in this manner is called multi- stage cluster sampling. FIGURE 12.5 The procedure for selecting a stratified sample

FIGURE 12.6 The concept of cluster sampling Non-random/non-probability sampling designs in quantitative research Non-probability sampling designs do not follow the theory of probability in the choice of elements from the sampling population. Non-probability sampling designs are used when the number of elements in a population is either unknown or cannot be individually identified. In such situations the selection of elements is dependent upon other considerations. There are five commonly used non- random designs, each based on a different consideration, which are commonly used in both qualitative and quantitative research. These are: 1. quota sampling; 2. accidental sampling; 3. judgemental sampling or purposive sampling; 4. expert sampling; 5. snowball sampling. What differentiates these designs being treated as quantitative or qualitative is the predetermined sample size. In quantitative research you use these designs to select a predetermined number of cases (sample size), whereas in qualitative research you do not decide the number of respondents in advance but continue to select additional cases till you reach the data saturation point. In addition, in qualitative research, you will predominantly use judgemental and accidental sampling strategies to select your respondents. Expert sampling is very similar to judgemental sampling except that in expert sampling the sampling population comprises experts in the field of enquiry. You can also use quota and snowball sampling in qualitative research but without having a predetermined number of cases in mind (sample size). Quota sampling The main consideration directing quota sampling is the researcher’s ease of access to the sample population. In addition to convenience, you are guided by some visible characteristic, such as gender or race, of the study population that is of interest to you. The sample is selected from a location convenient to you as a researcher, and whenever a person with this visible relevant characteristic is seen that person is asked to participate in the study. The process continues until you have been able to contact the required number of respondents (quota). Let us suppose that you want to select a sample of 20 male students in order to find out the average age of the male students in your class. You decide to stand at the entrance to the classroom, as this is convenient, and whenever a male student enters the classroom, you ask his age. This process continues until you have asked 20 students their age. Alternatively, you might want to find out about the attitudes of Aboriginal and Torres Strait Islander students towards the facilities provided to them in your university. You might stand at a convenient location and, whenever you see such a student, collect the required information through whatever method of data collection (such as interviewing, questionnaire) you have adopted for the study.

The advantages of using this design are: it is the least expensive way of selecting a sample; you do not need any information, such as a sampling frame, the total number of elements, their location, or other information about the sampling population; and it guarantees the inclusion of the type of people you need. The disadvantages are: as the resulting sample is not a probability one, the findings cannot be generalised to the total sampling population; and the most accessible individuals might have characteristics that are unique to them and hence might not be truly representative of the total sampling population. You can make your sample more representative of your study population by selecting it from various locations where people of interest to you are likely to be available. Accidental sampling Accidental sampling is also based upon convenience in accessing the sampling population. Whereas quota sampling attempts to include people possessing an obvious/visible characteristic, accidental sampling makes no such attempt. You stop collecting data when you reach the required number of respondents you decided to have in your sample. This method of sampling is common among market research and newspaper reporters. It has more or less the same advantages and disadvantages as quota sampling but, in addition, as you are not guided by any obvious characteristics, some people contacted may not have the required information. Judgemental or purposive sampling The primary consideration in purposive sampling is your judgement as to who can provide the best information to achieve the objectives of your study. You as a researcher only go to those people who in your opinion are likely to have the required information and be willing to share it with you. This type of sampling is extremely useful when you want to construct a historical reality, describe a phenomenon or develop something about which only a little is known. This sampling strategy is more common in qualitative research, but when you use it in quantitative research you select a predetermined number of people who, in your judgement, are best positioned to provide you the needed information for your study. Expert sampling The only difference between judgemental sampling and expert sampling is that in the case of the former it is entirely your judgement as to the ability of the respondents to contribute to the study. But in the case of expert sampling, your respondents must be known experts in the field of interest to you. This is again used in both types of research but more so in qualitative research studies. When you use it in qualitative research, the number of people you talk to is dependent upon the data saturation point whereas in quantitative research you decide on the number of experts to be contacted without considering the saturation point. You first identify persons with demonstrated or known expertise in an area of interest to you, seek their consent for participation, and then collect the information either individually or collectively in the form of a group.

FIGURE 12.7 Snowball sampling Snowball sampling Snowball sampling is the process of selecting a sample using networks. To start with, a few individuals in a group or organisation are selected and the required information is collected from them. They are then asked to identify other people in the group or organisation, and the people selected by them become a part of the sample. Information is collected from them, and then these people are asked to identify other members of the group and, in turn, those identified become the basis of further data collection (Figure 12.7). This process is continued until the required number or a saturation point has been reached, in terms of the information being sought. This sampling technique is useful if you know little about the group or organisation you wish to study, as you need only to make contact with a few individuals, who can then direct you to the other members of the group. This method of selecting a sample is useful for studying communication patterns, decision making or diffusion of knowledge within a group. There are disadvantages to this technique, however. The choice of the entire sample rests upon the choice of individuals at the first stage. If they belong to a particular faction or have strong biases, the study may be biased. Also, it is difficult to use this technique when the sample becomes fairly large. Systematic sampling design: a ‘mixed’ design Systematic sampling has been classified as a ‘mixed’ sampling design because it has the characteristics of both random and non-random sampling designs. In systematic sampling the sampling frame is first divided into a number of segments called intervals. Then, from the first interval, using the SRS technique, one element is selected. The selection of subsequent elements from other intervals is dependent upon the order of the element selected in the first interval. If in the first interval it is the fifth element, the fifth element of each subsequent interval will be chosen. Notice that from the first interval the choice of an element is on a random basis, but the choice of the elements from subsequent intervals is dependent upon the choice from the first, and hence cannot be classified as a random sample. The procedure used in systematic sampling is presented in Figure 12.8.

FIGURE 12.8 The procedure for selecting a systematic sample Although the general procedure for selecting a sample by the systematic sampling technique is described above, you can deviate from it by selecting a different element from each interval with the SRS technique. By adopting this, systematic sampling can be classified under probability sampling designs. To select a random sample you must have a sampling frame (Figure 12.9). Sometimes this is impossible, or obtaining one may be too expensive. However, in real life there are situations where a kind of sampling frame exists, for example records of clients in an agency, enrolment lists of students in a school or university, electoral lists of people living in an area, or records of the staff employed in an organisation. All these can be used as a sampling frame to select a sample with the systematic sampling technique. This convenience of having a ‘ready-made’ sampling frame may be at a price: in some cases it may not truly be a random listing. Mostly these lists are in alphabetical order, based upon a number assigned to a case, or arranged in a way that is convenient to the users of the records. If the ‘width of an interval’ is large, say, 1 in 30 cases, and if the cases are arranged in alphabetical order, you could preclude some whose surnames start with the same letter or some adjoining letter may not be included at all. Suppose there are 50 students in a class and you want to select 10 students using the systematic sampling technique. The first step is to determine the width of the interval (50/10 = 5). This means that from every five you need to select one element. Using the SRS technique, from the first interval (1–5 elements), select one of the elements. Suppose you selected the third element. From the rest of the intervals you would select every third element. The calculation of sample size Students and others often ask: ‘How big a sample should I select?’, ‘What should be my sample size?’ and ‘How many cases do I need?’ Basically, it depends on what you want to do with the findings and what type of relationships you want to establish. Your purpose in undertaking research is the main determinant of the level of accuracy required in the results, and this level of accuracy is an important determinant of sample size. However, in qualitative research, as the main focus is to explore or describe a situation, issue, process or phenomenon, the question of sample size is less important. You usually collect data till you think you have reached saturation point in terms of discovering new information. Once you think you are not getting much new data from your respondents, you stop collecting further information. Of course, the diversity or heterogeneity in what you are trying to find out about plays an important role in how fast you will reach saturation point. And remember: the greater the heterogeneity or diversity in what you are trying to find out about, the greater the number of respondents you need to contact to reach saturation point. In determining the size of your sample for quantitative studies and in particular for cause-and-effect studies, you need to consider the following:

FIGURE 12.9 Systematic sampling At what level of confidence do you want to test your results, findings or hypotheses? With what degree of accuracy do you wish to estimate the population parameters? What is the estimated level of variation (standard deviation), with respect to the main variable you are studying, in the study population? Answering these questions is necessary regardless of whether you intend to determine the sample size yourself or have an expert do it for you. The size of the sample is important for testing a hypothesis or establishing an association, but for other studies the general rule is: the larger the sample size, the more accurate your estimates. In practice, your budget determines the size of your sample. Your skills in selecting a sample, within the constraints of your budget, lie in the way you select your elements so that they effectively and adequately represent your sampling population. To illustrate this procedure let us take the example of a class. Suppose you want to find out the average age of the students within an accuracy of 0.5 of a year; that is, you can tolerate an error of half a year on either side of the true average age. Let us also assume that you want to find the average age within half a year of accuracy at the 95 per cent confidence level; that is, you want to be 95 per cent confident about your findings. The formula (from statistics) for determining the confidence limits is where = estimated value of the population mean = average age calculated from the sample t0.05 = value of t at 95 per cent confidence level σ/√η = standard error σ = standard deviation η = sample size

√ = square root If we decide to tolerate an error of half a year, that means *t-value from the following table There is only one unknown quantity in the above equation, that is σ. Now the main problem is to find the value of σ without having to collect data. This is the biggest problem in estimating the sample size. Because of this it is important to know as much as possible about the study population. The value of σ can be found by one of the following: 1. guessing; 2. consulting an expert; 3. obtaining the value of σ from previous comparable studies; or 4. carrying out a pilot study to calculate the value. Let us assume that σ is 1 year. Then Hence, to determine the average age of the class at a level of 95 per cent accuracy (assuming σ = 1 year) with half a year of error, a sample of at least 16 students is necessary. Now assume that, instead of 95 per cent, you want to be 99 per cent confident about the estimated age, tolerating an error of half a year. Then Hence, if you want to be 99 per cent confident and are willing to tolerate an error of half a year,

you need to select a sample of 27 students. Similarly, you can calculate the sample size with varying values of σ. Remember the golden rule: the greater is the sample size, the more accurately your findings will reflect the ‘true’ picture. Sampling in qualitative research As the main aim in qualitative enquiries is to explore the diversity, sample size and sampling strategy do not play a significant role in the selection of a sample. If selected carefully, diversity can be extensively and accurately described on the basis of information obtained even from one individual. All non-probability sampling designs – purposive, judgemental, expert, accidental and snowball – can also be used in qualitative research with two differences: 1. In quantitative studies you collect information from a predetermined number of people but, in qualitative research, you do not have a sample size in mind. Data collection based upon a predetermined sample size and the saturation point distinguishes their use in quantitative and qualitative research. 2. In quantitative research you are guided by your desire to select a random sample, whereas in qualitative research you are guided by your judgement as to who is likely to provide you with the ‘best’ information. The concept of saturation point in qualitative research As you already know, in qualitative research data is usually collected to a point where you are not getting new information or it is negligible – the data saturation point. This stage determines the sample size. It is important for you to keep in mind that the concept of data saturation point is highly subjective. It is you who are collecting the data and decide when you have attained the saturation point in your data collection. How soon you reach the saturation point depends upon how diverse is the situation or phenomenon that you are studying. The greater the diversity, the greater the number of people from whom you need to collect the information to reach the saturation point. The concept of saturation point is more applicable to situations where you are collecting information on a one-to-one basis. Where the information is collected in a collective format such as focus groups, community forums or panel discussions, you strive to gather as diverse and as much information as possible. When no new information is emerging it is assumed that you have reached the saturation point. Summary In this chapter you have learnt about sampling, the process of selecting a few elements from a sampling population. Sampling, in a way, is a trade-off between accuracy and resources. Through sampling you make an estimate about the information of interest. You do not find the true population mean. Two opposing philosophies underpin the selection of sampling units in quantitative and qualitative research. In quantitative studies

a sample is supposed to be selected in such a way that it represents the study population, which is achieved through randomisation. However, the selection of a sample in qualitative research is guided by your judgement as to who is likely to provide you with complete and diverse information. This is a non-random process. Sample size does not occupy a significant place in qualitative research and it is determined by the data saturation point while collecting data instead of being fixed in advance. In quantitative research, sampling is guided by three principles, one of which is that the greater the sample size, the more accurate the estimate of the true population mean, given that everything else remains the same. The inferences drawn from a sample can be affected by both the size of the sample and the extent of variation in the sampling population. Sampling designs can be classified as random/probability sampling designs, non-random/non-probability sampling designs and ‘mixed’ sampling designs. For a sample to be called a random sample, each element in the study population must have an equal and independent chance of selection. Three random designs were discussed: simple random sampling, stratified random sampling and cluster sampling. The procedures for selecting a sample using these designs were detailed step by step. The use of the fishbowl technique, the table of random numbers and specifically designed computer programs are three commonly used methods of selecting a probability sample. There are five non-probability sampling designs: quota, accidental, judgemental, expert and snowball. Each is used for a different purpose and in different situations in both quantitative and qualitative studies. In quantitative studies their application is underpinned by the sample size whereas the data saturation point determines the ‘sample size’ in qualitative studies. Systematic sampling is classified under the ‘mixed’ category as it has the properties of both probability and non-probability sampling designs. The last section of the chapter described determinants of, and procedures for, calculating sample size. Although it might be slightly more difficult for the beginner, this was included to make you aware of the determinants involved as questions relating to this area are so commonly asked. In qualitative research, the question of sample size is less important, as your aim is to explore, not quantify, the extent of variation for which you are guided by reaching saturation point in terms of new findings. For You to Think About Refamiliarise yourself with the keywords listed at the beginning of this chapter and if you are uncertain about the meaning or application of any of them revisit these in the chapter before moving on. Consider the implications of selecting a sample based upon your choice as a researcher and how you could make sure that you do not introduce bias. In the absence of a sampling frame for employees of a large organisation, which sampling design would you use to select a sample of 219 people? Explain why you would choose this design and the process you would undertake to ensure that the sample is representative. From your own area of interest, identify examples of where cluster sampling could be applied. What determines sample size in qualitative research? What is the data saturation point in qualitative studies?

STEP V Writing a Research Proposal This operational step includes one chapter: Chapter 13: Writing a research proposal



CHAPTER 13 How to Write a Research Proposal In this chapter you will learn about: The purpose of a research proposal in quantitative and qualitative research How to structure a research proposal How to write a research proposal Keywords: conceptual framework, data analysis, data processing, hypothesis, limitations, literature review, research design, research problem, sampling, study design, study objectives, theoretical framework, time-frame. The research proposal in quantitative and qualitative research All research endeavours, in both qualitative and quantitative research, in every academic and professional field are preceded by a research proposal. It informs your academic supervisor or potential research contract provider about your conceptualisation of the total research process that you propose to undertake so that they can examine its validity and appropriateness. In any academic field, your research proposal will go through a number of committees for approval. Unless it is approved by all of them, you will not be able to start your research. Hence, it is important for you to study closely what constitutes a research proposal. You need to write a research proposal whether your research study is quantitative or qualitative and in both cases you use a similar structure. The main difference is in the proposed procedures and methodologies for undertaking the research endeavour. When providing details for different parts of the research proposal, for quantitative studies, you will detail quantitative methods, procedures and models and, for qualitative studies, your proposed process will be based upon methods and procedures that form the qualitative research methodology. Certain requirements for a research proposal may vary from university to university, and from discipline to discipline within a university. What is outlined here will satisfy most requirements but you should be selective regarding what is needed in your situation. A research proposal is an overall plan, scheme, structure and strategy designed to obtain answers to the research questions or problems that constitute your research project. A research proposal

should outline the various tasks you plan to undertake to fulfil your research objectives, test hypotheses (if any) or obtain answers to your research questions. It should also state your reasons for undertaking the study. Broadly, a research proposal’s main function is to detail the operational plan for obtaining answers to your research questions. In doing so it ensures and reassures the reader of the validity of the methodology for obtaining answers to your research questions accurately and objectively. In order to achieve this function, a research proposal must tell you, your research supervisor and reviewers the following information about your study: what you are proposing to do; how you plan to find answers to what you are proposing; why you selected the proposed strategies of investigation. Contents of a research proposal A research proposal should contain the following information about your study: an introduction, including a brief literature review; theoretical framework that underpins your study; conceptual framework which constitutes the basis of your study; objectives or research questions of your study; hypotheses to be tested, if applicable; study design that you are proposing to adopt; setting for your study; research instrument(s) you are planning to use; sampling design and sample size; ethical issues involved and how you propose to deal with them; data processing procedures; proposed chapters of the report; problems and limitations of the study; proposed time-frame for the project. A research proposal should communicate the above contents clearly and specifically in such a way that anyone going through it should be able to undertake all tasks in the same manner as you would have. It should also: enable you to return to the proposal for your own guidance in decision making at different stages of the research process; convince your research supervisor or a reviewer that your proposed methodology is meritorious, valid, appropriate and workable in terms of obtaining answers to your research questions or objectives. Universities and other institutions may have differing requirements regarding the style and content

of a research proposal. Requirements may also vary within an institution, from discipline to discipline or from supervisor to supervisor. (The guidelines set out in this chapter therefore provide a framework within which a research proposal should be written.) Your proposal should follow the suggested guidelines and be written in an academic style. It must contain appropriate references in the body of the text and a bibliography at the end. Your survey of the relevant literature should cover major publications on the topic. The theoretical framework for your study must emerge from this literature review and must have its grounding in empirical evidence. As a rule, the literature review includes: a conceptual framework, and theoretical and empirical information about the main issues under study; some of the major research findings relating to your topic, research questions raised in the literature and gaps identified by previous researchers. Your literature review should also raise issues relating to the methodology you are proposing. For example, it may examine how other studies operationalised the major variables of relevance to your study and may include a critique of methodology relevant to your study. The critiques of methods and procedures should be included under their respective headings. For example, a critique of the sampling design you adopt should be included under ‘sampling’ or a critique to the study design should be discussed under ‘study design’. Note that the suggested research proposal structure does not contain a section entitled ‘survey of the literature’ or ‘literature review’. This is because references to the literature should be integrated with your arguments conceptually rather than chronologically and should become a part of all the aspects of your research report from problem conceptualisation to conclusions. The literature should be reviewed under main themes that emerge from your reading of the literature and should be included in the ‘introduction’ and ‘the problem’. Issues identified in the literature to do with research methodology and problems pertinent to the various aspects of research procedures should be discussed under their respective headings. For example, issues pertaining to the study design under ‘study design’, issues relating to sampling under ‘sampling’ and the literature pertaining to the research instrument under the ‘measurement procedure’. In suggesting this format it is assumed that you are reasonably well acquainted with research methodology and an academic style of writing. That is, you know how to write a set of objectives or construct a hypothesis, you are familiar with the various study designs and you can construct a research instrument and cite a reference. The pages that follow outline a framework for a research proposal. The contents under each heading may vary markedly from discipline to discipline, according to the academic level of the student (BA Hons, MA, PhD) and whether your study is predominantly quantitative or qualitative. For quantitative proposals you need to be very specific in proposing how you are going to undertake each step of the research journey, whereas for qualitative research proposals such details are not expected as your methodology is flexible and unstructured to accommodate in-depth search. However, you need to provide a broad approach to your enquiry as a part of your research proposal. Each section of the proposed outline for a research proposal is divided into two parts: 1. a suggested title for the section and an outline of its contents;

2. examples outlining contents for the section – the same four examples of research projects, each taken from a different discipline, are used as illustrations in each section. Preamble/introduction The proposal should start with an introduction to include some of the information listed below. Remember that some of the contents suggested in this section may not be relevant to certain studies, so use your discretion in selecting only what is pertinent to your study. In writing this section, the literature review (see Chapter 3 on reviewing the literature) is of central importance as it serves two main functions: 1. It acquaints you with the available literature in the area of your study, thereby broadening your knowledge base. 2. It provides you with information on the methods and procedures other people have used in similar situations and tells you what works and what does not. The type, extent and quality of a literature review are mostly dependent upon the academic level for which you are writing the proposal. The contents of this section may also vary greatly according to the subject area under study. Start with a very broad perspective of the main subject area, before gradually narrowing the focus to the central problem under investigation. In doing so, cover the following aspects of your study area: an overview of the main area under study; a historical perspective (development, growth, etc.) pertinent to the study area; philosophical or ideological issues relating to the topic; trends in terms of prevalence, if appropriate; major theories, if any; the main issues, problems and advances in the subject area under study; important theoretical and practical issues relating to the central problem under study; the main findings relating to the core issue(s). Four examples of possible topics for the preamble/introduction for a research proposal follow. Example A Suppose that you are conducting a study to investigate the impact of immigration on the family. The preamble/introduction should include a brief description of the following: The origins of migratory movements in the world. General theories developed to explain migratory behaviour. The reasons for migration. Current trends in migration (national and state).

The impact of immigration on family roles and relationships (e.g. on husband and wife, on children and parents, on parental expectations of children, etc.). Occupational mobility. etc. Example B Suppose your research project is to conduct a study of the attitudes of foster carers towards foster payment in … (name of the place/state/country). The preamble/introduction would include the following: The origins of foster placement, the philosophy of foster care, a historical overview of foster care and changes over the years. Reasons for foster care and changes over time. The origins of foster placement in … (the country in which you are conducting your study). The effects of foster placement on children and parents. Policies with respect to foster care in … (the region). The origins of foster care in … (the region). Administrative procedures for foster care in … (the region). The training of foster parents in … (the region). The role and responsibility of foster parents. etc. Example C Suppose that you plan to study the relationship between academic achievement and social environment. The preamble/introduction would include the following: The role of education in our society. Major changes in the philosophy of education over time. Factors affecting attitudes towards education. The development of education in … (country). Trends in education participation rates in … (country) with particular reference to the region in which the study is being carried out. Changing educational values. Role of parents and peers in academic achievement. Impact of social environment on academic achievement. etc.

Example D Suppose you are undertaking a qualitative study to find out what it means to have a child with ADHD in the family. The preamble/introduction should include your thoughts and arguments, and what the literature says around the following aspects of ADHD. Definitions and symptoms of ADHD. Causes of ADHD. Medical perspective on ADHD. Effects of ADHD on family life. Treatment for ADHD. Implications for a child if untreated. Management of ADHD. etc. The problem Having provided a broad introduction to the area under study, now focus on issues relating to its central theme, identifying some of the gaps in the existing body of knowledge. Identify some of the main unanswered questions. Here some of the main research questions that you would like to answer through your study should also be raised, and a rationale and relevance for each should be provided. Knowledge gained from other studies and the literature about the issues you are proposing to investigate should be an integral part of this section. Specifically, this section should: identify the issues that are the basis of your study; specify the various aspects of/perspectives on these issues; identify the main gaps in the existing body of knowledge; raise some of the main research questions that you want to answer through your study; identify what knowledge is available concerning your questions, specifying the differences of opinion in the literature regarding these questions if differences exist; develop a rationale for your study with particular reference to how your study will fill the identified gaps. The following examples outline the topics about which the literature should be reviewed and included in the section entitled ‘The problem’. Keep in mind that these are just suggestions and should serve only as examples for you to develop and change as you feel appropriate for your own study. Example A What settlement process does a family go through after immigration? What adjustments do immigrants have to make?

What types of change can occur in family members’ attitudes? (Theory of acculturation etc.) What is the possible impact of settlement on family roles and relationships? In terms of impact, what specific questions do you want to answer through the study? What does the literature say about these questions? What are the different viewpoints on these issues? What are your own ideas about these questions? What do you think will be the relevance of the findings of your study to the existing body of knowledge and to your profession? How will the findings add to the body of knowledge and be useful to professionals in your field? etc. Example B What are the broad issues, debates, arguments and counter-arguments regarding foster-care payment? What are the attitudes of foster parents to the amount, mode and form of payment and what does the literature say about these issues? What are the different viewpoints/perspectives regarding payment for foster care? What main questions will your study answer? How will your findings help in policy formulation and programme development? etc. Example C What theories have been developed to explain the relationship between academic achievement and social environment? What is the relationship between educational achievement and social environment: what theoretical model will be the basis of your study? What do previous theories and researches have to say regarding the components of the theoretical model and academic achievement? For example, the relationship between academic achievement and: — the self-esteem and aspirations/motivation of a student; — peer group influence; — parental involvement and its relationship with their socioeconomic status; — the motivation and interest of students in the subject; — employment prospects; — relationship with a teacher; — etc.

Example D What are the effects on the family of having a child with ADHD in the family as identified in the literature? According to the literature, are there any differences between these effects and the type of family? What strategies have been used for the management of ADHD by a family? What effects, according to the literature, does ADHD have on sibling relationships? What are the perceptions of family members about the effects and management of ADHD? How do families cope when they have a child with ADHD in the family? etc. Objectives of the study In this section include a statement of both your study’s main and subobjectives (see Chapter 4). Your main objective indicates the central thrust of your study whereas the subobjectives identify the specific issues you propose to examine. The objectives of the study should be clearly stated and specific in nature. Each subobjective should delineate only one issue. Use action-oriented verbs such as ‘to determine’, ‘to find out’ and ‘to ascertain’ in formulating subobjectives, which should be numerically listed. If the objective is to test a hypothesis, you must follow the convention of hypothesis formulation in wording the specific objectives. In qualitative studies the statement of objectives is not as precise as in quantitative studies. In qualitative studies you should simply mention an overall objective of the study as your aim is to explore as much as possible as you go along. As you know, the strength of qualitative research is in flexibility of approach and the ability to incorporate new ideas while collecting data. Having structured statements that bind you to a predetermined framework of exploration is not a preferred convention in qualitative research. Statements like to explore ‘what does it mean to have a child with ADHD in the family?’, ‘how does it feel to be a victim of domestic violence?’, ‘how do people cope with racial discrimination?’, ‘the relationship between resilience and yoga’ or ‘reconstructing life after bushfire’, are sufficient to communicate your intent of objectives in qualitative research. More detailed objectives, if need be, can be developed after a study is complete. Example A Main objective: To ascertain the impact of immigration on the family. Subobjectives:

1. To determine the impact of immigration on husband/wife roles as perceived by immigrants. 2. To find out the impact of immigration on marital relations. 3. To ascertain perceived changes in parental expectations of children’s academic and professional achievement. 4. To determine perceived changes of attitude towards marriage in the study population. Example B Main objective: To determine the opinion of foster carers about the form and extent of foster payment they feel they should receive for taking care of a foster child. Subobjectives: 1. To determine the form and mode of payment for taking care of a foster child. 2. To identify the factors that foster parents believe should be the basis for determining the rate of payment for fostering a child. 3. To determine the relationship, if any, between the socioeconomic graphic characteristics of foster parents and their views on payment. Example C Main objective: To examine the relationship between academic achievement and social environment. Subobjectives: 1. To find out the relationship, if any, between self-esteem and a student’s academic achievement at school. 2. To ascertain the association between parental involvement in a student’s studies and his/her academic achievement at school. 3. To examine the links between a student’s peer group and academic achievement. 4. To explore the relationship between academic achievement and the attitude of a student towards teachers. Example D Main objective: To explore what it means to have a child with ADHD in the family.

Hypotheses to be tested A hypothesis is a statement of your assumptions about the prevalence of a phenomenon or about a relationship between two variables that you plan to test within the framework of the study (see Chapter 6). If you are going to test hypotheses, list them in this section. When formulating a hypothesis you have an obligation to draw conclusions about it in the text of the report. Hypotheses have a particular style of formulation. You must be acquainted with the correct way of wording them. In a study you may have as many hypotheses as you want to test. However, it is not essential to have a hypothesis in order to undertake a study – you can conduct a perfectly satisfactory study without formulating a hypothesis. Example A H1 = In most cases there will be a change in husband/wife roles after immigration. H2 = In a majority of cases there will be a change in parents’ expectations of their children. Hi = etc. Example B H1 = Most people become foster parents because of their love of children. H2 = A majority of foster parents would like to be trained to care for foster children. Hi = etc. Example C H1 = A student’s self-esteem and academic achievement at school are positively correlated. H2 = The greater the parental involvement in a student’s studies, the higher the academic achievement. H3 = A student’s attitude towards teachers is positively correlated with his/her academic achievement in that subject. Hi = etc. Example D Hypotheses are not constructed in qualitative research. Study design

Describe the study design (for details see Chapter 8) you plan to use to answer your research questions. (For example, say whether it is a case study, descriptive, cross-sectional, before-and-after, experimental or non-experimental design.) Identify the strengths and weaknesses of your study design. Include details about the various logistical procedures you intend to follow while executing the study design. One characteristic of a good study design is that it explains the details with such clarity that, if someone else wants to follow the proposed procedure, s/he will be able to do exactly as you would have done. Your study design should include information about the following: Who makes up the study population? Can each element of the study population be identified? If yes, how? Will a sample or the total population be studied? How will you get in touch with the selected sample? How will the sample’s consent to participate in the study be sought? How will the data be collected (e.g. by interview, questionnaire or observation)? In the case of a mailed questionnaire, to what address should the questionnaire be returned? Are you planning to send a reminder regarding the return of questionnaires? How will confidentiality be preserved? How and where can respondents contact you if they have queries? Example A The study is primarily designed to find out from a cross-section of immigrants from …, … and … (names of the countries) the perceived impact of immigration on family roles. Initial contact with the ethnic associations for these countries will be made through the elected office bearers to obtain a list of members. Five immigrants will be selected from the list at random, and will be contacted by phone to explain the purpose of the study and its relevance, and to seek their agreement to participate in the study. Those who give their consent will be interviewed at their homes or any other convenient place. To select a further sample, a snowball sampling technique will be used until the desired sample size is obtained. Example B The study design is cross-sectional in nature, being designed to find out from a cross-section of foster parents their opinions about foster payment. All foster parents currently registered with the Department of … (name of the office) constitute the study population. From the existing records of this department it seems that there are 457 foster parents in … (name of the region). As it is impossible for the researcher, within the constraints of time and money, to collect information from all the foster parents, it is proposed to select a sample of 50 per cent of the study population with the proposed sampling strategy. The questionnaire, with a supporting letter from the department will be sent with a prepaid envelope. The respondents will be requested to return the questionnaire by … (date). The letter from the researcher attached to the questionnaire will explain the objectives and relevance of the study, assure the respondents of anonymity and give them the option of not participating in the study if they wish. A contact number will be provided in case a respondent has any questions. In the case of a low response rate (less than 25 per cent), a reminder will be sent to respondents. Example C It is proposed that the study will be carried out in two government high schools in the metropolitan area. The principals of the schools most accessible to the researcher will be contacted to explain the purpose of the study and the help needed from the

school, and to seek their permission for the students to participate in the study. As the constraints of time and resources do not permit the researcher to select more than two schools, negotiations with other schools will cease when two schools agree to participate in the study. It is proposed to select Year 9 students as the academic achievement of students in Years 8 and 10 could be affected by factors unique to them. Year 8 students may be experiencing anxiety as a result of having just made the transition to a new system. The motivation of students in Year 10 could be affected by their being at the stage in their education where they must decide if they will stay on at school. In order to control the variance attributable to the gender of a student it is proposed to select only male students. Once the principal of a school agrees to allow the study to be carried out, the researcher will brief the teacher in charge about the study and its relevance, and will arrange a date and time for administering the questionnaire. When the students are assembled, ready to participate in the study, the researcher will explain its purpose and relevance, and then distribute the questionnaire. The researcher will remain with the class to answer any questions the students might have. Example D The researcher is known to a family that has a child with ADHD and that belongs to an ADHD support group which meets every month. The researcher proposes to make initial contact with the group through the known family. The researcher will attend one of the monthly meetings and brief the group on the purpose and relevance of the study, criteria for inclusion in the study, what it entails to be involved in the study, and other aspects of the study. The respondents will also be assured of the anonymity of the information shared by them and its ethical use. The members of the group will be encouraged to ask questions about any aspect of the study. Having sought their consent, the researcher will seek opinions of some group members to decide who should participate in the study in light of the inclusion criteria. It is proposed to select six families, three where both parents are involved in the treatment and management of an ADHD child and three from families where the mother is the sole carer. This is primarily to see if there are differences in looking after a child with ADHD among different types of family. The potential respondents will be individually contacted by the researcher to seek their consent for participation in the study. Once consent has been obtained the place and timings for interviews will be fixed with each family. Depending upon the type of family, the issues will be discussed either collectively with the father and mother or with the mother only. Before starting an interview their permission to record the interview on a tape recorder will be sought. Having completed the interviews, the researcher will transcribe the responses and a copy will be given to the respondents for confirmation and validation. The setting Briefly describe the organisation, agency or community in which you will conduct your study. If the study is about a group of people, highlight some of the salient characteristics of the group (e.g. its history, size, composition and structure) and draw attention to any available relevant information. If your research concerns an agency, office or organisation, include the following in your description: the main services provided by the agency, office or organisation; its administrative structure; the type of clients served; information about the issues that are central to your research. If you are studying a community, briefly describe some of the main characteristics, such as: the size of the community;

a brief social profile of the community (i.e. the composition of the various groups within it); issues of relevance to the central theme of your study. Note that, due to the nature of the content, it would be difficult to provide examples. Measurement procedures This section should contain a discussion of your instrument (see Chapters 9 and 10) and the details of how you plan to operationalise your major variables (Chapter 5). To start with, justify your choice of research tool, highlighting its strengths and pointing out its weaknesses. Then outline the major segments of your research tool and their relevance to the main objectives of the study. If you are using a standard instrument, briefly discuss the availability of evidence on its reliability and validity. If you adapt or modify it in any way, describe and explain the changes you have made. You should also discuss how you are going to operationalise the major concepts. For example, if measuring effectiveness, specify how it will be measured. If you plan to measure the self-esteem of a group of people, mention the main indicators of self-esteem and the procedures for its measurement (e.g. the Likert or Thurstone scale, or any other procedure). Ideally, for quantitative studies you should attach a copy of the research instrument to your proposal. Note that, due to the nature of the content, it would be difficult to provide examples for this section. Ethical issues All academic institutions are particular about any ethical issues that research may have. To deal with them, all institutions have some form of policy on ethics. You need to be acquainted with your institution’s policy. It is imperative that in your proposal you identify any ethical issues and describe how you propose to deal with them. You need to look at the ethical issues particularly from the viewpoint of your respondents and, in case of any potential ‘harm’, psychological or otherwise, you need to detail the mechanism in place to deal with it. Further information on ethical issues is provided in Chapter 14. Sampling Under this section of the proposal include the following (consult Chapter 12 on sampling): the size of the sampling population (if known) and from where and how this information will be obtained; the size of the sample you are planning to select and your reasons for choosing this size; an explanation of the sampling design you are planning to use in the selection of the sample (simple random sampling, stratified random sampling, quota sampling, etc.).

Example A Because a lack of information as to the exact location of migrant families makes it difficult to use a probability sampling design, it is proposed that the researcher will employ a snowball sampling technique. The researcher will make initial contact with five families who have emigrated from … (name of the country) during the past seven to ten years, who are either known to him/her or on the basis of information obtained from the office bearers of the formal associations representing the migrant groups. From each respondent the researcher will obtain names and addresses of other immigrants who have come from the same country during the same period. The respondents thus identified will then be interviewed and asked to identify other respondents for the researcher. This process will continue until the researcher has interviewed 70 respondents. Example B Because of the constraints of time and resources it is proposed to select 50 per cent of the foster parents currently registered (457) with the department using the systematic random sampling technique. Every other foster parent registered with the department will be selected, thus 229 individuals will constitute the sample for the study. Example C The selection of schools will be done primarily through quota sampling. Schools will be selected on the basis of their geographical proximity to the researcher. The researcher will prepare a list of schools, in rank order, of accessibility. Once two schools agree to participate in the study, negotiations with other schools will cease. All Year 9 male students will constitute the study population. It is expected that the sample will not exceed 100 students. Example D It is proposed to use the judgemental/purposive sampling technique to select six families from the group, three where both parents look after an ADHD child and three where only the mother has the main responsibility (single parent families). On the basis of informal discussions with the group members, those families who are expected to be information rich in treating and managing a child with ADHD will be selected to be interviewed. Analysis of data In general terms, describe the strategy you intend to use for data analysis (Chapter 15). Specify whether the data will be analysed manually or by computer. For computer analysis, identify the program and where appropriate the statistical procedures you plan to perform on the data. For quantitative studies also identify the main variables for cross-tabulation. For qualitative studies, describe how you plan to analyse your interviews or observation notes to draw meanings from what your respondents have said about issues discussed or observation notes made. One of the common techniques is to identify main themes, through analysing the contents of the information gathered by you in the field. You first need to decide whether you want to analyse this information manually or use a computer program for the purpose. There are three ways to proceed with content analysis:

1. From your field notes develop a framework of your write-up and as you go through your notes directly integrate that information within the structure developed. If you adopt this method, you need to be reasonably clear about the structure. It does not mean that you cannot develop the structure as you go on analysing; still, a clear vision will be of immense help in slotting information gathered in the field by you into the write-up. 2. The second method is that you transcribe your field notes to be read by you over and over again to identify the main themes. These themes become the basis of your write-up. 3. There are computer programs such as NUD*IST, Ethnograph, NVivo specifically designed to handle descriptive data. You may prefer to use one of these programs. These programs are also based upon the principle of content analysis. The only difference is that instead of your searching manually, they identify where a particular text identifying the theme appears. You need to specify which particular strategy you are proposing for data analysis for your study. Example A Frequency distributions in terms of: age; education; occupation; number of children; duration of immigration; etc. Cross-tabulations: Impact of husband/wife roles age; number of children; education; occupation; etc. Example B Frequency distributions in terms of: age; income; education;

occupation; marital status; duration of foster care; number of foster children; etc. Cross-tabulations: Attitude towards foster payment age; number of children; education; occupation; etc. Statistical tests to be applied: chi square; regression analysis; etc. Example C Frequency distributions in terms of: age; parents’ occupation; parents’ educational levels; students’ occupational aspirations; parental involvement in students’ studies; self-esteem; peer group influence; number of hours spent on studies; etc. Cross-tabulations: Academic achievement peer group influence; parental involvement in students’ studies; self-esteem; occupational aspirations;

attitude towards teachers; etc. Example D The in-depth interviews carried out with the families will be transcribed using Microsoft Word. These transcribed interviews will be closely studied to identify the main themes they communicate. These themes will be sorted by issues relating to management and treatment of a child with ADHD. The themes will then become part of the write-up. Structure of the report As clearly as possible, state how you intend to organise the final report (see Chapter 17). In organising your material for the report, the specific objectives of your study are of immense help. Plan to develop your chapters around the main themes of your study. The title of each chapter should clearly communicate the main thrust of its contents. The first chapter, possibly entitled ‘Introduction’, should be an overall introduction to your study, covering most of your project proposal and pointing out deviations, if any, from the original plan. The second chapter should provide some information about the study population itself – that is, some of its socioeconomic–demographic characteristics. The main aim of this chapter is to give readers some background on the population from which you collected the information. The second chapter, therefore, may be entitled, ‘Socioeconomic–demographic characteristics of the study population’ or ‘The study population’ or any other title that communicates this theme to readers. Titles for the rest of the chapters will vary from study to study but, as mentioned, each chapter should be written around a main theme. Although the wording of chapter titles is an individual choice, each must communicate the main theme of the chapter. In developing these themes the specific objectives of the study should be kept in the front of your mind. If your study is qualitative, the main issues identified during data collection and analysis stages should become the basis of developing chapters. Having developed significant issues, the next step is to organise the main themes under each issue and develop a structure that you will follow to communicate your findings to your readers. Example A It is proposed that the report will be divided into the following chapters: Chapter 1: Introduction The socioeconomic–demographic characteristics of the study Chapter 2: population The impact on husband/wife roles Chapter 3: The impact on marital relations Chapter 4: The impact on expectations of children Chapter 5:

Chapter 6: The impact on attitudes towards marriage Chapter 7: Summary, conclusions and recommendations Example B The dissertation will be divided into the following chapters: Chapter 1: Introduction Chapter 2: A profile of the study population Chapter 3: Foster carers’ perceptions of their role Chapter 4: Attitudes of foster carers towards foster-care payment Chapter 5: The preferred method of payment Chapter 6: General comments made by respondents about foster care Chapter 7: Summary, conclusions and recommendations Example C The report will have the following chapters: Chapter 1: Introduction Chapter 2: The study population Chapter 3: Occupational aspirations, self-esteem and academic achievement Chapter 4: The extent of parental involvement and academic achievement Chapter 5: Peer group influence and academic achievement Chapter 6: Academic achievement and student attitudes towards teachers Chapter 7: Summary, conclusions and recommendations Example D It is proposed that the report will have the following chapters: Chapter 1: ADHD: A theoretical perspective Issues and difficulties faced by family members in bringing up a child Chapter 2: with ADHD ADHD and its perceived effects on the child Chapter 3: ADHD and its perceived impact on sibling relationships Chapter 4: Managing treatment Chapter 5: Perceived effects of ADHD on schooling of the child Chapter 6: Perceived effects of ADHD on relationships with other children Chapter 7:

Chapter 8: A case history Chapter 9: Summary and conclusions Problems and limitations This section should list any problems you think you might encounter concerning, for example, the availability of data, securing permission from the agency/organisation to carry out the study, obtaining the sample, or any other aspect of the study. You will not have unlimited resources and as this may be primarily an academic exercise, you might have to do less than an ideal job. However, it is important to be aware of – and communicate – any limitations that could affect the validity of your conclusions and generalisations. Here, problems refer to difficulties relating to logistical details, whereas limitations designate structural problems relating to methodological aspects of the study. In your opinion the study design you chose may not be the best but you might have had to adopt it for a number of reasons. This is classified as a limitation of the study. This is also true for sampling or measurement procedures. Such limitations should be communicated to readers. Appendix As an appendix, in the case of quantitative studies, attach your research instrument. Also, attach a list of references in the appendix of the proposal. Work schedule You must set yourself dates as you need to complete the research within a certain time-frame. List the various operational steps you need to undertake and indicate against each the date by which you aim to complete that task. Remember to keep some time towards the end as a ‘cushion’ in case the research process does not go as smoothly as planned. Develop a chart as shown in Table 13.1. TABLE 13.1 Developing a time-frame for your study

Summary A research proposal details the operational plan for obtaining answers to research questions. It must tell your supervisor and others what you propose to do, how you plan to proceed and why the chosen strategy has been selected. It thus assures readers of the validity of the methodology used to obtain answers accurately and objectively. The guidelines set out in this chapter provide only a framework within which a research proposal for both quantitative and qualitative studies should be written and assume that you are reasonably well acquainted with research methodology and an academic style of writing. The contents of your proposal are arranged under the following headings: preamble/introduction, the problem, objectives of the study, hypotheses to be tested, study design, setting, measurement procedures, sampling, analysis of data, structure of the report, and problems and limitations. The specifics, under each heading, will vary with the type of study you are proposing to undertake. The write-up for qualitative studies will be based upon qualitative methodology and quantitative methodology will determine the contents of quantitative studies. The ‘preamble’ or ‘introduction’ introduces the main area of the study. To start with, the literature review is broad and then it gradually narrows to the specific problem you are investigating. The theoretical framework should be a part of this section. The next section, ‘the problem’, details the specific problem under study. The research questions for which you are planning to find answers are raised in this section. ‘Objectives of the study’ contains your main objectives and your subobjectives. Hypotheses, if any, should be listed in the section ‘hypotheses to be tested’. The logistical procedures you intend to follow are detailed under ‘study design’. ‘The setting’ consists of a description of the organisation or community in which you plan to conduct your study. The procedure for obtaining information and the measurement of major variables are explained in the ‘measurement procedures’ section. You need to write about ethical issues that your study might have and how you propose to deal with them. How you will select your sample is described under ‘sampling’. The procedure for data analysis is discussed under ‘analysis of data’. The way you plan to structure your report is outlined under ‘structure of the report’. Anticipated problems in conducting the study and limitations with its design are described under ‘problems and limitations’. As an appendix to your proposal attach a copy of the research instrument and a list of the references. A work schedule provides a time-frame for your study. For You to Think About Refamiliarise yourself with the keywords listed at the beginning of this chapter and if you are uncertain about the meaning or application of any of them revisit these in the chapter before moving on. Compare the research proposal contents suggested in this chapter with those recommended by your university or department. If they are different, what are the differences? Find out the process that a research proposal goes through in your university before approval is granted.

STEP VI Collecting Data This operational step includes one chapter to make you aware of the ethical issues in research: Chapter 14: Considering ethical issues in data collection



CHAPTER 14 Considering Ethical Issues in Data Collection In this chapter you will learn about: Ethics: the concept Stakeholders in research Ethical issues to consider concerning research participants Ethical issues to consider relating to the researcher Ethical issues to consider regarding the sponsoring organisation Keywords: bias, code of conduct, confidentiality, deprivation of treatment, ethos, harm, informed consent, principles of conduct, research participants, sensitive information, sponsoring organisations, stakeholders, subjectivity. Ethics: the concept All professions are guided by a code of ethics that has evolved over the years to accommodate the changing ethos, values, needs and expectations of those who hold a stake in the professions. Some professions are more advanced than others in terms of the level of development of their code of ethics. Some have very strict guidelines, monitor conduct effectively and take appropriate steps against those who do not abide by the guidelines. Most professions have an overall code of conduct that also governs the way they carry out research. In addition, many research bodies have evolved a code of ethics separately for research. Medicine, epidemiology, business, law, education, psychology and other social sciences have well- established codes of ethics for research. Let us first examine what we mean by ‘ethics’ or ‘ethical behaviour’. According to the Collins Dictionary (1979: 502), ethical means ‘in accordance with principles of conduct that are considered correct, especially those of a given profession or group’. The keywords here, ‘principles of conduct’ and ‘considered correct’, raise certain questions: What are these principles of conduct?

Who determines them? In whose judgement must they be considered correct? Closely related questions are as follows: Are there universal principles of conduct that can be applied to all professions? Do these change with time? Should they? What happens when a professional does not abide by them? The subject of ethics needs to be considered in light of these questions. The way each profession serves society is continuously changing in accordance with society’s needs and expectations and with the technology available for the delivery of a service. The ethical codes governing the manner in which a service is delivered also need to change. What has been considered ethical in the past may not be so judged at present, and what is ethical now may not remain so in the future. Any judgement about whether a particular practice is ethical is made on the basis of the code of conduct prevalent at that point in time. As the service and its manner of delivery differ from profession to profession, no code of conduct can be uniformly applied across all professions. Each profession has its own code of ethics, though there are commonalities. If you want guidelines on ethical conduct for a particular profession, you need to consult the code of ethics adopted by that profession or discipline. ‘What are these principles of conduct?’ is the most important question as it addresses the issue of the contents of ethical practice in a profession. As the code of conduct varies from profession to profession, it is not possible to provide a universal answer to this question. However, in research, any dilemma stemming from a moral quandary is a basis of ethical conduct. There are certain behaviours in research – such as causing harm to individuals, breaching confidentiality, using information improperly and introducing bias – that are considered unethical in any profession. The next question is: in whose judgement must a code of conduct be considered correct? Who decides whether a particular practice is wrong? If a procedure is carried out wrongly, what penalties should be imposed? It is the overall body of professionals or government organisations that collectively develops a professional code of conduct and forms a judgement as to whether or not it is being followed. As mentioned, most professions have established an overall code of ethics and also a code of ethics for conducting research in their respective fields. As this book is designed for researchers in the social sciences, we will examine ethical issues relating to research in general and issues that are applicable to most social science disciplines. Stakeholders in research There are many stakeholders in research, whether it is quantitative or qualitative. It is important to look at ethical issues in relation to each of them. The various stakeholders in a research activity are: 1. the research participants or subjects;

2. the researcher; 3. the funding body. Who should be considered as a research participant varies from profession to profession. Generally, all those with direct and indirect involvement in a research study are considered as research participants, hence stakeholders. In addition, those who are likely to be affected by the findings of a study are also considered as stakeholders. In the fields of medicine, public health, epidemiology and nursing, patients and non-patients who become part of a study and those who participate in an experiment to test the effectiveness of a drug or treatment are considered as research participants. Service providers, service managers and planners who are involved in either providing the service or collecting information relating to the study are also stakeholders in the research. In the social sciences, the participants include individuals, groups and communities providing information to help a researcher to gain understanding of a phenomenon, situation, issue or interaction. In social work and psychology, participants include clients as well as non-clients of an agency from whom information is collected to find out the magnitude of a problem, the needs of a community or the effectiveness of an intervention; and service providers, social workers and psychologists, when they provide information for a study. In marketing, consumers as well as non-consumers of a product provide information about consumption patterns and behaviour. In education, subjects include students, teachers and perhaps the community at large who participate in educational research activities. Similarly, in any discipline in which a research activity is undertaken, those from whom information is collected or those who are studied by a researcher become participants of the study. Researchers constitute the second category of stakeholders. Anyone who collects information for the specific purpose of understanding, consolidation, enhancement and development of professional knowledge, adhering to the accepted code of conduct, is a researcher. S/he may represent any academic discipline. Funding organisations responsible for financing a research activity fall into the third category of stakeholders. Most research is carried out using funds provided by business organisations, pharmaceutical companies, service institutions (government, semi-government or voluntary), research bodies and/or academic institutions. The funds are given for specific purposes. Each category of stakeholders in a research activity may have different interests, perspectives, purposes, aims and motivations that could affect the way in which the research activity is carried out and the way results are communicated and used. Because of this, it is important to ensure that research is not affected by the self-interest of any party and is not carried out in a way that harms any party. It is therefore important to examine ethical conduct in research concerning different stakeholders under separate categories. Ethical issues to consider concerning research participants There are many ethical issues to consider in relation to the participants of a research activity. Collecting information One could ask: why should a respondent give any information to a researcher? What right does a

researcher have to knock at someone’s door or to send out a questionnaire? Is it ethical to disturb an individual, even if you ask permission before asking questions? Why should a person give you his/her time? Your request for information may create anxiety or put pressure on a respondent. Is this ethical? But the above questions display a naive attitude. The author believes that if this attitude had been adopted, there would have been no progress in the world. Research is required in order to improve conditions. Provided any piece of research is likely to help society directly or indirectly, it is acceptable to ask questions, if you first obtain the respondents’ informed consent. Before you begin collecting information, you must consider the relevance and usefulness of the research you are undertaking and be able to convince others of this also. If you cannot justify the relevance of the research you are conducting, you are wasting your respondents’ time, which is unethical. Seeking consent In every discipline it is considered unethical to collect information without the knowledge of participants, and their expressed willingness and informed consent. Seeking informed consent ‘is probably the most common method in medical and social research’ (Bailey 1978: 384). Informed consent implies that subjects are made adequately aware of the type of information you want from them, why the information is being sought, what purpose it will be put to, how they are expected to participate in the study, and how it will directly or indirectly affect them. It is important that the consent should also be voluntary and without pressure of any kind. Schinke and Gilchrist write: Under standards set by the National Commission for the Protection of Human Subjects, all informed-consent procedures must meet three criteria: participants must be competent to give consent; sufficient information must be provided to allow for a reasoned decision; and consent must be voluntary and uncoerced. (1993: 83) Competency, according to Schinke and Gilchrist, ‘is concerned with the legal and mental capacities of participants to give permission’ (1993: 83). For example, some very old people, those suffering from conditions that exclude them from making informed decisions, people in crisis, people who cannot speak the language in which the research is being carried out, people who are dependent upon you for a service and children are not considered to be competent. Providing incentives Is it ethical to provide incentives to respondents to share information with you? Some researchers provide incentives to participants for their participation in a study, feeling this to be quite proper as participants are giving their time. Others think that the offering of inducements is unethical. In the author’s experience most people do not participate in a study because of incentives, but because they realise the importance of the study. Therefore, giving a small gift after having obtained your information, as a token of appreciation, is in the author’s opinion not unethical. However, giving a present before data collection is unethical. Seeking sensitive information

Information sought can pose an ethical dilemma in research. Certain types of information can be regarded as sensitive or confidential by some people and thus an invasion of privacy. Asking for this information may upset or embarrass a respondent. However, if you do not ask for the information, it may not be possible to pursue your interest in the area and contribute to the existing body of knowledge. For most people, questions on sexual behaviour, drug use and shoplifting are intrusive. Even questions on marital status, income and age may be considered to be an invasion of privacy by some. In collecting data you need to be careful about the sensitivities of your respondents. The dilemma you face as a researcher is whether you should ask sensitive and intrusive questions. In the author’s opinion it is not unethical to ask such questions provided that you clearly and frankly tell your respondents the type of information you are going to ask, and give them sufficient time to decide if they want to share the information with you, without any major inducement. The possibility of causing harm to participants Is the research going to harm participants in any way? Harm includes: not only hazardous medical experiments but also any social research that might involve such things as discomfort, anxiety, harassment, invasion of privacy, or demeaning or dehumanising procedures. (Bailey 1978: 384) When you collect data from respondents or involve subjects in an experiment, you need to examine carefully whether their involvement is likely to harm them in any way. If it is, you must make sure that the risk is minimal. Minimum risk means that the extent of harm or discomfort in the research is not greater than that ordinarily encountered in daily life. It is unethical if the way you seek information creates anxiety or harassment, and if you think it may happen, you need to take steps to prevent this. Maintaining confidentiality Sharing information about a respondent with others for purposes other than research is unethical. Sometimes you need to identify your study population to put your findings into context. In such a situation you need to make sure that at least the information provided by respondents is kept anonymous. It is unethical to identify an individual respondent and the information provided by him/her. Therefore, you need to ensure that after the information has been collected, its source cannot be identified. In certain types of study you might need to visit respondents repeatedly, in which case you will have to identify them until the completion of your visits. In such situations you need to be extra careful that others do not have access to the information. It is unethical to be negligent in not protecting the confidentiality and anonymity of the information gathered from your respondents. If you are doing research for someone else, you need to make sure that confidentiality is maintained by this party as well. Ethical issues to consider relating to the researcher

Avoiding bias Bias on the part of the researcher is unethical. Bias is different from subjectivity. Subjectivity, as mentioned earlier, is related to your educational background, training and competence in research, and your philosophical perspective. Bias is a deliberate attempt either to hide what you have found in your study, or to highlight something disproportionately to its true existence. It is absolutely unethical to introduce bias into a research activity. If you are unable to control your bias, you should not be engaging in the research. Remember, it is the bias that is unethical and not the subjectivity. Provision or deprivation of a treatment Both the provision and deprivation of a treatment may pose an ethical dilemma for you as a researcher. When testing an intervention or a treatment, a researcher usually adopts a control experiment design. In such studies, is it ethical to provide a study population with an intervention or treatment that has not yet been conclusively proven effective or beneficial? But if you do not test a treatment/intervention, how can you prove or disprove its effectiveness or benefits? On the other hand, you are providing an intervention that may not be effective. Is this ethical? Is it ethical to deprive the control group of a treatment even if it may prove to be only slightly effective? And beyond the issue of control groups, is it ethical to deprive people who are struggling for life of the possible benefit, however small, which may be derived from a drug that is only under trial? As a researcher you need to be aware of these ethical issues. There are arguments and counter-arguments about these issues. However, it is usually accepted that deprivation of a trial treatment to a control group is not unethical as, in the absence of this, a study can never establish the effectiveness of a treatment which may deprive many others of its possible benefits. This deprivation of the possible benefits, on the other hand, is considered by some as unethical. There are no simple answers to these dilemmas. Ensuring informed consent, ‘minimum risk’ and frank discussion as to the implications of participation in the study may help to resolve some of these ethical issues. Using inappropriate research methodology A researcher has an obligation to use appropriate methodology, within his/her knowledge base, in conducting a study. It is unethical to use deliberately a method or procedure you know to be inappropriate to prove or disprove something that you want to, such as by selecting a highly biased sample, using an invalid instrument or by drawing wrong conclusions. Incorrect reporting To report the findings in a way that changes or slants them to serve your own or someone else’s interest is unethical. Correct and unbiased reporting of the findings are important characteristics of ethical research practice. Inappropriate use of the information

How will the information obtained from respondents be used by the researcher? The use of information in a way that directly or indirectly affects respondents adversely is unethical. Can information be used adversely to affect the study population? If so, how can the study population be protected? As a researcher you need to consider and resolve these issues. Sometimes it is possible to harm individuals in the process of achieving benefits for organisations. An example would be a study to examine the feasibility of restructuring an organisation. Restructuring may be beneficial to the organisation as a whole but may be harmful to some individuals. Should you ask respondents for information that is likely to be used against them? If you do, the information may be used against them, and if you do not, the organisation may not be able to derive the benefits of restructuring. In the author’s opinion, it is ethical to ask questions provided you tell respondents of the potential use of the information, including the possibility of its being used against some of them, and you let them decide if they want to participate. Some may participate for the betterment of the organisation even though it may harm them and others may decide against it. However, to identify either of them is unethical in research. Ethical issues regarding the sponsoring organisation Restrictions imposed by the sponsoring organisation Most research in the social sciences is carried out using funds provided by sponsoring organisations for a specific purpose. The funds may be given to develop a programme or evaluate it; to examine its effectiveness and efficiency; to study the impact of a policy; to test a product; to study the behaviour of a group or community; or to study a phenomenon, issue or attitude. Sometimes there may be direct or indirect controls exercised by sponsoring organisations. They may select the methodology, prohibit the publication of ‘what was found’ or impose other restrictions on the research that may stand in the way of obtaining and disseminating accurate information. Both the imposition and acceptance of these controls and restrictions are unethical, as they constitute interference and could amount to the sponsoring organisation tailoring research findings to meet its vested interests. The misuse of information How is the sponsoring body going to use the information? How is this likely to affect the study population? Sometimes sponsoring organisations use research as a pretext for obtaining management’s agenda. It is unethical to let your research be used as a reason for justifying management decisions when the research findings do not support them. However, it is recognised that it may be extremely difficult or even impossible for a researcher to prevent this from happening. Summary This chapter is designed to make you aware of the ethical issues to be considered when conducting research. The ethical issues to be considered are the same in both quantitative and qualitative research. How you resolve them depends upon you, and the